- From: Jonathan Robie <jwrobie@mindspring.com>
- Date: Mon, 16 Feb 2004 12:57:36 -0500
- To: Michael Rys <mrys@microsoft.com>
- Cc: public-qt-comments@w3.org
I agree that we are not likely to change the schemas in our examples - but we should encourage our users to write the kinds of queries that will survive schema changes that do not affect the logic of the query. Jonathan Michael Rys wrote: >But we are not going to change the schemata in our examples. // has a >couple of issues, especially in the context of static type inference >that makes it not the best choice to formulate many queries and often >cannot be optimized better than the path expressions. > >Best regards >Michael > > > >>-----Original Message----- >>From: Jonathan Robie [mailto:jwrobie@mindspring.com] >>Sent: Monday, February 16, 2004 9:53 AM >>To: Michael Rys >>Cc: public-qt-comments@w3.org >>Subject: Re: [XQuery] MS-XQ-LC1-110 >> >>I disagree. In general, queries using // are more robust to changes in >>the schema than queries that spell out the entire path, and are a >> >> >better > > >>way to write such queries, as long as the underlying implementation >>optimizes appropriately. >> >>Jonathan >> >>Michael Rys wrote: >> >> >> >>>General Comment >>>Examples >>> >>>We should reduce the amount of examples that use // and instead write >>>them with the full path. This is giving a bad example to users in >>> >>> >cases > > >>>where a clear path exists. >>> >>> >>> >>> >>> > > > >
Received on Monday, 16 February 2004 13:00:10 UTC